Max Robotics — Global Service Robot Directory + US Certification

About

Harry Zhang is a roboticist whose research focuses on dynamic manipulation, grasp planning, and self-supervised learning for robotic systems. His most influential work, "Robots of the Lost Arc" (2021, 38 citations), introduces a self-supervised learning framework that enables a UR5 robot to perform high-speed dynamic manipulation of fixed-endpoint cables—vaulting over obstacles, knocking objects from pedestals, and weaving between them—demonstrating how robots can master complex, agile tasks without human supervision. Zhang also contributed to accessible grasp planning through "Dex-Net AR" (2020, 5 citations), which leverages commodity smartphones and augmented reality to democratize robotic grasping, and "6-DoF Grasp Planning using Fast 3D Reconstruction and Grasp Quality CNN" (2020, 7 citations), which reduces reliance on expensive depth cameras. His work on "TAX-Pose" (2022, 4 citations) addresses task-specific cross-pose estimation for manipulating unseen objects, while "AVPLUG" (2021, 3 citations) tackles mechanical search in cluttered environments. Most recently, Zhang has ventured into space robotics with "POLAR-Sim" (2025, 2 citations), augmenting NASA's lunar dataset for data-driven perception and rover simulation. With a growing citation footprint and a knack for combining practical hardware with clever learning algorithms, Zhang is advancing the frontier of dexterous, autonomous robot manipulation.